diy solar

diy solar

16cell 48v Eve Grade A battery, one strange cell V jumps when charge current present

12kw_2021

New Member
Joined
May 3, 2021
Messages
162
Hello

I have JK bms with 2A balance current controlling 16 Eve 280Ah cells that are configured into a 48V battery.
Issues I have is it appears that the BMS cycle count has jumped and is inaccurate (In use < 100 days, but cycle count is > 300).
One cell #14 jumps from 3.3V to 3.8V which then triggers charge shutdown for overvoltage, but it appears that the battery is not fully charged.

No.. I did not top balance this set of cells, as my understanding is that over many charge and discharge cycles the large 2A balance current available from the JK BMS will effectively execute a top Balance . (Andy in Australia says so for what it is worth.).

I am attaching some images from the JK BMS control screen, that shows that the "sense" wires all have low/suitable resistance, so I don't think the sense wires are the issue.

The batteries have the latest style top posts with two screws and the cell to cell bus bars have been tightened using a torque wrench.

Do I assume that the constant V jump of Cell 14 that disables the BMS due to cell overvoltage, appears to prevent the charger from fully charging the battery to be defective and replace it, or are there any other investigations of the unexpected Cell 14 behaviour, as well as the impossible BMS cycle count.

Note that as soon as charge is manually disabled cell 14 drops and is very well matched to the other cells (Note that all 16 cells are within .001 V with charge turned off, but as soon as charge is turned back on cell 14 V jumps up to 3.6-3.8 V and triggers the cell overvoltage protection.

See images of this Battery with low and higher state of charge, I don't believe I am getting anything close to the expected battery capacity from this battery, I am trying to establish battery capacity available and will update when I have that information.

I also have a hard time believing the JK BMS 98% charge level as correct with the reported voltages in the attached.

Thanks in advance
 

Attachments

  • cell 14 investigation.pdf
    337 KB · Views: 20
Hello

I have JK bms with 2A balance current controlling 16 Eve 280Ah cells that are configured into a 48V battery.
Issues I have is it appears that the BMS cycle count has jumped and is inaccurate (In use < 100 days, but cycle count is > 300).
One cell #14 jumps from 3.3V to 3.8V which then triggers charge shutdown for overvoltage, but it appears that the battery is not fully charged.

No.. I did not top balance this set of cells, as my understanding is that over many charge and discharge cycles the large 2A balance current available from the JK BMS will effectively execute a top Balance . (Andy in Australia says so for what it is worth.).

Well, it does appear the 2A balancer doesn't top balance that well, does it? And why does it allow a cell to reach 3.8V before high cell voltage disconnect when it should be 3.65V?

I'm all for initial charge when top balancing using a BMS, no problem there. But I also advocate then putting cells in parallel and using a power supply and charging to 3.65V. It doesn't take that long.

I've watched Andy before, given a choice, I prefer watching something else if the opportunity is there to learn something. His videos go on and on. Nothing against Andy, just that sometimes brevity can be a moving experience, I will only live X number of hours in my lifetime, I prefer to use them wisely. Do I think Andy is the gospel on battery cells? No.


I am attaching some images from the JK BMS control screen, that shows that the "sense" wires all have low/suitable resistance, so I don't think the sense wires are the issue.

The batteries have the latest style top posts with two screws and the cell to cell bus bars have been tightened using a torque wrench.

Do I assume that the constant V jump of Cell 14 that disables the BMS due to cell overvoltage, appears to prevent the charger from fully charging the battery to be defective and replace it, or are there any other investigations of the unexpected Cell 14 behaviour, as well as the impossible BMS cycle count.

The cycle count is most likely due to the disconnects. The cell behavior is questionable, I see in the pdf that at rest the cell voltage is lower.

However, without an actual top balance in parallel, it can not be clearly determined. One other thing, that cell most likely is damaged due to the 3.8V charge.

I would also swap that cell position on the BMS to ensure the BMS is not faulty for the #14 slot balancing and monitoring. As you have to disassemble the bank, might as well top balance in parallel.

Note that as soon as charge is manually disabled cell 14 drops and is very well matched to the other cells (Note that all 16 cells are within .001 V with charge turned off, but as soon as charge is turned back on cell 14 V jumps up to 3.6-3.8 V and triggers the cell overvoltage protection.

See images of this Battery with low and higher state of charge, I don't believe I am getting anything close to the expected battery capacity from this battery, I am trying to establish battery capacity available and will update when I have that information.

I also have a hard time believing the JK BMS 98% charge level as correct with the reported voltages in the attached.

Thanks in advance
 
Yes I would double check your connections and move the cell somewhere else. Also double check your BMS settings and make sure everything is set right, for example balancing needs to turn on at around 3.35 to 3.4 volts during charging.
 
I have a similar problem with one of my hand crafted with love 16s packs. I use a JBD BMS so no active balancer, just the one built in. I’ve top balanced this one pack too many times to admit trying to resolve this. Same problematic battery in the pack no matter where I move it in the series. Anyway what works for me now very well is stating the JDB balancer at lower voltage, 3325mv. This keeps it from running ahead of the rest of the cells and they all get to full charge together. I saw that once that battery hit 3395mv it would take off and hit 3.65v much faster than the rest.

This is probably not the best solution but it’s the one working for me at this point in time.
 
Just an update, I have two batteries , and disabled the charge on the one that is operating correctly to then use the charger to find out how much power the battery that read zero SOC before the sun came up. After a very short period of charging it now reads 98% charged, and it did not get nearly enough charge to get there. I also see one Cell 14 overvoltage event in the log so far this AM, so that troublesome cell is still an issue. I fully agree that I need to move that cell to a different location. What is the best way to identify that cell in my layout.

Confused on why the JK BMS would go from 0% SOC to 98% SOC with nearly no charging. Guess I need to find out how to reset the BMS to force it to recalibrate itself.

I did not fully understand the explanation as to why the cycle count on this battery would be so high, can anyone help me on that?

I have two identical 200A JK BMS, and I may also swap them to see if that changes anything, and yes I will go over all my settings in the BMS.

Also note that the 4.8V readings observed that cause charge shutdown are very short , so I am hoping that short duration is not causing damage to cell 14.

My layout is below

Am i counting correctly to #14 in my layout below?

And yes Andy is very long winded, but if you play his stuff on 1.5x you can glean some tidbits, but no, I don't watch every vid he posts.

1680874191206.png
Thanks for the replies.. looks like I may be ordering a replacement cell from the 18650 battery store.
 
Cells are counted in order from negative which is #1 and not positive post which is #16.

#14 would be opposite your diagram.
 
Back
Top